Settlement of a marine tube worm as a function of current velocity: Interacting effects of hydrodynamics and behavior I

Settlement cxpcriments were conducted with larvae of Phragmatopoma lapidosa callfirnica (a reefbuilding sabcllariid polychaetc) in turbulent flume flows (near-surface velocities of 5, 10, 15, 20, 25, 30, and 35 cm s-l) over a hydrodynamically smooth bed. Boundary shear velocities spanned the critical shear velocity for initiation of particle motion and for suspended-load transport of passive larval mimics. Larvae were allowed one pass over a scdimcnt array with two treatments: tube sand, a natural inducer of metamorphosis, and noninductive sand. Delivery of larvae to the array was the result of interactions between the flow regime and larval behavior. At intermediate flows (15, 20, and 25 cm s-l), where numbers of metamorphosed juveniles and total animals (larvae + juveniles) in the array were maximal, larvae tumbled along the flume bottom, as did the passive larval mimics. At slower llows, larvae actively left the bottom and swam into the water, passing over the array. At the fastest flows, hydrodynamics alone may have reduced settlement because larvae, like the mimics, were eroded from the bed and carried as suspended load over the array or because enhanced turbulent mixing distributed larvae more evenly in the water, thus reducing their concentration close to the bed. Once delivered to the substratum, behavioral responses to chemical cues were ultimately responsible for metamorphosis; in all flows, > 96% of metamorphosed juveniles were in the tube sand, whereas most unmetamorphosed larvae were in noninductive sand. Explanations for the observed patterns of distribution and abundance of benthic marine organisms have shifted in the past dccade’from those invoking postrccruitment mortality (disturbance predation, competition) to those in’ Dcdicatcd to the memory of Douglas P. Wilson, D-SC., F.I. Biol., Hon. F.R.P.S., 1902-1991. Acknowledgments We thank V. R. Starczak for assistance with cxperimental design, statistical analyses, and comments on the manuscript, J. H. Trowbridge for calculating boundary shear stress from the velocity profiles, and three anonymous reviewers for improving the manuscript. Our conceptual picture of the behavior of relatively large, lowdensity particles in steady, turbulent flows was improved by discussions with J. H. Trowbridge, W. B. Dade, and P. S. Hill. We acknowledge the Coastal Research Center at the Woods Hole Oceanographic Institution for USC of the flume facilities. Financial support for this research was provided by NSF grant OCE 881265 1 to C. A. Butman, ONR grant NO0